Araştırma Makalesi

Tobit modeline bayesci yaklaşım ve winBUGS ile uygulama

Cilt: 12 Sayı: 1 15 Ocak 2022
PDF İndir
EN TR

Bayesian approach to tobit model and application with winBUGS

Abstract

In this study, Tobit Model which is known to censored regression model is examined. The aim of the study is to estimate the Tobit model parameters using with Bayesian approach and compare the Bayesian and classical estimation results. At first, the methods are examined which are used to estimate the Tobit model parameters. These methods are the maximum likelihood method and the Bayesian estimation method. For the inferences of the Bayesian approach, the Gibbs sampling method is used. In the application section first, the simulation program was written in the Matlab program for the Tobit model which is censored from below, and the classical maximum likelihood method is used to estimate the parameters of the Tobit model. Later, Bayesian estimations are obtained by importing the same data which are generated using the Matlab simulation program, from the WinBugs program. And then the estimation results are obtained from using the data of Accord, Mazda 6, and Maxima models, which are the most preferred automobile brands in the USA in 2017. The estimation results are similar for the two methods. Besides, classical confidence intervals and Bayesian credible intervals are computed. It is seen that Bayesian highest density intervals are slightly narrower than the classical confidence intervals.

Keywords

Bayesian approach , MCMC (Markov Chain Monte Carlo) , Qualitative variable , Tobit model , WinBUGS

Kaynakça

  1. Abbas, H. K. and Thaher, R. J. (2019). Bayesian adaptive lasso tobit regression. Journal of AL-Qadisiyah for Computer Science and Mathematics, 11(1), 1-10, https://doi.org/10.29304/jqcm.2019.11.1.471.
  2. Adarabioyo, I. and Awe, O. O. (2020). Application of bayesian tobit regression to global radiation, 2020 International Conference in Mathematics, Computer Engineering and Computer Science (ICMCECS). https://doi.org/10.1109/ICMCECS47690.2020.240848.
  3. Alhamzawi, R. and Ali, H.T.M. (2018). Bayesian tobit quantile regression with L_(1/2) penalty, Communications in Statistics -Simulation and Computation, 47(6), 1739- 1750. https://doi.org/10.1080/03610918.2017.1323224.
  4. Austin, P. C., Escobar, M. and Kopec, A. J. (2000). The use of tobit model for analyzing measures of health status. Ouality of Life Research, 901-910.
  5. Austin, P. C. (2002). Bayesian extensions of the tobit model for analyzing measures of health status. Medical Decision Making, 22(2), 152-162. https://doi.org/10.1177/0272989X0202200212.
  6. Baba, V.V. (1990). Methodological issue in modeling absence: a comparison of least squares and tobit analyses. Journal of Applied Psychology. 75(4), 428–432. https://doi.org/10.1037/0021-9010.75.4.428.
  7. Bilir, K.B.Ö. (2016).Bayesyen markov zinciri monte carlo simülasyonu. Doktora Tezi, Uludağ Üniversitesi, Sosyal Bilimler Enstitüsü, Bursa.
  8. Chib, S. (1992). Bayes inference in the tobit censored regression model, Journal of Econometrics, 51(1-2), 79-99. https://doi.org/10.1016/0304-4076(92)90030-U.
  9. Dagenais, G.M. (1975). Application of a threshold regression model to household purchases of automobiles. The Review of Economics and Statistics, 57(3), 275-285.
  10. Dagne, G. and Huang, Y.(2012). Bayesian inference for a nonlinear mixed-effects tobit model with multivariate skew-T distributions: application to AIDS studies. Int J Biostat, 8(1). https://doi.org/10.1515/1557-4679.1387.

Kaynak Göster

APA
Kaya, S., & Köksal Babacan, E. (2022). Tobit modeline bayesci yaklaşım ve winBUGS ile uygulama. Gümüşhane Üniversitesi Fen Bilimleri Dergisi, 12(1), 187-199. https://doi.org/10.17714/gumusfenbil.873998